Early Stage Structural Estimation

[TDG]

Hey everyone! I'm new here, but I am working on a design project and have a few questions, mostly surrounding structure. I have done quite a bit of design work, but I'm not as familiar with composite work on small boats.

Project background: My project is a 12 m trimaran (LOA == LWL), with a vaka/center hull that's about 2.2 m BWL, and amas/side hulls are about 0.6 m BWL with an overall beam of 10 m. My goal is ultimately to be an oceangoing-capable vessel and meeting all the appropriate ISO requirements for the same.

I'm doing the initial weight estimate, and I'm trying to figure out an appropriate structure. I have a few questions: (1) Effectiveness of sandwich panels. (2) Initial structural layup. (3) Necessity of longitudinal structure. (4) Aca/cross structure loads. (5) Ancillary structure. (6) Other thoughts.

1) For the vaka, I would like to do 75 mm/3" of PVC (H80 or similar) foam not only for structure, but also for insulation and flotation. When I studied composites back in the day, the core of a sandwich panel was simply the "web" with fairly limited structural effectiveness, but also no limit as far as as thickness. However, Larsson & Eliasson make a comment when discussing sandwich panel cores which suggests that there is a limit to their thickness due to a need to transfer the load between the inner and outer lamina. Is there a good rule of thumb for this? Is 3" too much? What about semi-hollow 4"? (For the weather deck, I'm running longitudinal conduit for both running rigging and distributed systems in the main deck/cabin overhead.)

2) The immersed vaka bottom structure is a stiffened panel design with (currently) both longitudinal and transverse beams, as well as 8 full height bulkheads (3 WT/FT). As mentioned, the vaka has 75 mm H80, and, based on the "Multihull Structure Thoughts" post on this forum, it seemed reasonable to assume a single layer of 1708 on each side of the core. (Assumptions: 50% wetout with PVE versus manufacturer recommended 50% by weight. Manufacturer recommended paint/coatings. 12.5% weight margin for all weight due to uncertainty.) How unreasonable is that? For the amas, I went down to 18 mm foam on the sides and deck with 1708, retaining ~75 mm foam only on the bottom (from the baseline up to the waterline). Is that too light? In addition to transverse bulkheads, there is a full length deck about mid-way up.

3) Looking at most production 7-15 m boats I've seen, there isn't much in the way of longitudinal structure. I understand that under about 50 m boats typically experience seaway loading as a global load, and so the shell structure is typically sufficient. I currently have a pretty decent amount of weight budgeted for 7 x 100 mm wide floor beams (1708 top/sides, H80 core - it would probably be more effective to use uniaxial glass rather than 1708 for that). How excessive is that? Just use one or two centerline floors to distribute the mast loading? just use localized beams under the mast?

4) For the larger ships I typically work on, we use class-based seaway loading for transverse loading of the acas. (We also use seaway loading for longitudinal strength, but that's another conversation.) However, much of the transverse strength calculations for sailboats are based on righting moment (e.g. rig strength). I currently have a way overstrength cross structure (a total of 10 carbon fibre beams with 2 m of vertical separation between the top 4 and the bottom 6 - one of the beams, on its own, could take the full force at beyond the maximum righting moment of the craft). Is there a better way to calculate transverse loading?

5) For ancillary structure (e.g. outfitting, non-structural weather-exposed panels, the coach roof, etc.), how light can I reasonably go? For surfaces that will have people on them (e.g. benches, the decks in the amas), I currently have 18 mm of H80 foam with either 1708 (the decks) or wood veneer (the benches). Is that too light? For the vertical surfaces, could I go down to H20 with 6 oz boat cloth (or veneer)? For the coach roof, I currently have 75 mm thick H20 with 6 oz boat cloth. Is that too light to resist a crashing wave, or just a human handling sails? (While the coach roof is over 4 m AWL, I have certainly been in situations where I've gotten wet much higher up.) What about near-vertical forward facing structure? Is that too light?

6) What else should I probably be thinking of?

Thanks,
- R/TDG

 

[oldmulti]

R/TDG. Please purchase some plans EG Scarab 32 ft trimaran ($150 australian or about $110 US) and the Joe dobler 30 foot tri plans from Duckworks for about $40 US. Both are electronic downloads. Both give valuable information on structures, internal fitouts, foils etc. Yes design your own tri but practical experience from designers help you understand the boundaries of what can be done. They are a cheap education.

 

[catsketcher]

A couple of points as I built a 12m trimaran - a Newick we switched from cedar to foam.

The foam seems way too thick. We used 12mm foam for the amas, 15mm for the main hull and 25 mm for the wing.

There is a huge variation in laminate weight. My cedar cat uses 600 db outside and 400 uni inside the hulls, but a friends similar style cat was about half that. Other friends have over twice that. Yet all boats are still around, some just get more dinged than others. But I would stay clear of anything lighter than 400 on the vertical surfaces and 600 for anything you walk on.

You should consider micro cracking. Shuttleworth talked about this. Basically composite light composite structure can start off stiff but then through repeated cycling and high impacts can soften - a la a Laser dinghy and the deck where you sit. So you need to have enough beef in the laminate to stay clear of fatigue loads.

Like Old multi says, when we shifted the Newick from cedar to foam, we bought a set of Farrier F 31 plans. It got a bit confusing when Ian rang me up to see how the tri was going. He had heard that we bought lots of foam and stuff and I had to explain that we were not building one of his boats. I would recommend you get some Farrier plans to look over. The laminates will be lighter than you want - unless you can find an F36 set. Look over and see how complicated a light laminate can get.

And another thing - we tend not to use DB (+- 45) with foam, we normally use biax (0-90)

cheers

Phil

 

[Rumars]

The only boats with a core that thick that I know of, were buildt by injecting PU foam between two premade skins. I don't even know if you can get 75mm PVC foam without a special order. Anyway since most of the foam is for insulation and flotation it is much cheaper to use a "normal" thickness sandwich, and glue PU or XPS to the inside. This has the added benefit that you could use this foam as a male mold, shape the flotation foam (by hand or CNC) into a plug, then laminate from the inside out (by hand or vacuum assisted). Where you will be installing the bulkheads (or other things bonded to the inside skin) you put some tape over the foam and cut that section out after laminating.

 

[TDG]

All,
Thanks for all of the thoughts! I'll look into some appropriate plans. I focused my comparative study on production boats (Dragonfly, Corsair, Neel), which was probably a mistake. I like Rumars's suggestion, since that will probably also help reduce weight. I really appreciate Phil's answers since I think that gets to a lot of what I was looking for.

Phil, I'm not too familiar with cedar strip construction. The one wooden ship I worked on utilized the glass purely for waterproofing. How much of your structure is in the cedar and how much in the glass?

Thanks,
- R/TDG

 

[redreuben]

PVC foam is an expensive product At 75mm your costs are going be at best 3x what they need to be. Even 20mm on this boat is overkill, however I appreciate the desire for built in flotation and perhaps if you have a need or desire for large unsupported panels then the 20mm is justified.
I would say 12-15 would be your ballpark figure.

 

[catsketcher]

As has been said above - stand on the shoulders of giants - there will be enough to worry about without reinventing laminates. My cat was 600gm double bias outside laminate, 12 mm thick cedar planks 40 ft long and about 40mm wide, then 400 gm uni inside.

Laminates are sort of a done deal. The main loads are going to be sitting on the beach with something hard under the boat, being pulled out for antifouling (all of the weight on a few points) the dinghy banging topsides and when bumping into marinas. Global sailing loads are all way lower. Bulkheads and interior furniture being well bonded to the hulls is really important but rarely gets talked about in reducing hull panel size. So does hull shape. The rounded amas of the Newick I built were very stiff with their shape, whereas a flat sided ama would have needed extra stiffening inside. My cat has a 2 metre long sort of bench that we use to step on to get into the bunk, but one vital job is to reduce panel size of unsupported interior hull.

A "normal" laminate is about 600 biax - 15mm foam - 400 biax for many foam hulls around 35ft. Go through Old Multis thread and you will see that the laminates go up and down. Whether the light ones stay stiff is one thing, but rarely do they break.

 

[TDG]

All,
Thanks so much! This is exactly what I was looking for. I'm not looking for a detailed structural design yet. Initially, I'm just working on a feasibility study to make sure it's reasonable. It sounds like I'm going way heavier than is strictly necessary - which is where I'd really like to be. I can jenny craig the boat and save on the order of probably half a ton. Once I get through this design cycle, I can go back and do a more detailed design to perform the final structural calculations.
Best,
- R/TDG